Why would a reward make a memory more likely to be strengthened overnight?
Sleep is not a neutral playback of the day. The hippocampus, a structure central to forming new episodic memories, is thought to reactivate recent experiences offline, and this replay is a leading candidate for how memories get consolidated into long-term storage [2]. Because the brain cannot keep everything, it has to triage: some experiences get replayed and strengthened, others are skipped. Framing this as an emergency-room triage is apt, since the system must decide which memories are worth the cost of consolidation [2].
Reward is one plausible priority signal. In one study, rewarded experiences were preferentially reactivated during early NREM sleep, and that reactivation was associated with better subsequent memory performance [1]. In plain terms, the sleeping brain revisited the rewarded material more, and the people who showed this pattern remembered it better. That is consistent with reward acting as a tag that makes a memory more likely to be replayed and strengthened, though the study supports an association rather than proving reward causes the strengthening.
Repetition and familiarity also decide which memories get replayed
Reward is not the only factor. In rats running repeated spatial tracks and then sleeping, the rate of sleep replay for a given track rose in proportion to how many times the animal had run it [2]. Repetition, in other words, raised a memory's priority for consolidation.
Familiarity pulled in the opposite direction. If a rat had already run the same track before its most recent sleep session, the rate of sleep replay for that track went down, as if the brain judged that a well-known route needed less further consolidation [2]. The same study found that the cumulative number of replay events during waking behavior, shaped by both novelty and how long the experience lasted, predicted which memories were prioritized for sleep replay [2]. This offers a simpler account of selective strengthening that does not require a reward signal: what gets replayed may reflect how much and how recently the experience was engaged during the day.
Is there direct evidence that the hippocampus and striatum replay together?
Not in these studies. The reward study observed preferential reactivation during early NREM sleep and better memory afterward, and its authors suggest this may implicate coordinated replay of hippocampal and striatal regions during sleep [1]. That is a hypothesis about mechanism, not a measurement of it. The striatum is a reward- and action-related region, so pairing it with the hippocampus is a natural guess for how motivation could bias consolidation, but the guess has not been directly tested here.
A study that did record coordinated activity during sleep points to a different partner. It examined hippocampal CA1 cells together with thalamic head-direction cells in mice during non-REM sleep and reported synchronized replay events, while finding no coordinated replay between the cell types in question [3]. That work concerns spatial signals rather than motivationally important memories, so it neither confirms nor rules out a hippocampal-striatal link. It does show that coordinated sleep replay can be measured directly, and that the identity of the replay partners matters.
So how should you read the claim that sleep strengthens motivated memories?
The behavioral side of the story is reasonably supported: rewarded experiences are reactivated more during early NREM sleep, and that reactivation is linked to better memory [1]. The triage logic is also supported in animals, where repetition raises and familiarity lowers the replay rate for a given experience [2]. What remains a proposal rather than a finding is the specific hippocampal-striatal mechanism. The most defensible summary is that sleep appears to favor motivationally important memories, and that coordinated hippocampal-striatal replay is a plausible but not yet directly demonstrated explanation for how that preference is implemented [1][3].
Sources used in this answer
Reward biases spontaneous neural reactivation during sleep
Rewarded experiences were preferentially reactivated during early NREM sleep, and this reactivation was associated with enhanced subsequent memory performance; the authors suggest this may implicate coordinated hippocampal-striatal replay but do not measure it.
The role of experience in prioritizing hippocampal replay
In rats, sleep replay rate for a spatial track increased with the number of times it was run and decreased with prior familiarity, and the cumulative amount of awake replay predicted which memories were prioritized for sleep replay.
Role of the head direction signal in memory replay
Hippocampal CA1 cells and thalamic head-direction cells showed synchronized replay during non-REM sleep in mice, while no coordinated replay was found between the cell types in question; the study did not address motivationally significant memories.